INDEX

Explanations

terms related to "mal" or "malignant" conditions

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 quæ

-1.09

 Grecs

-1.05

 itſelf

-1.03

 дописавши

-1.01

 chofe

-1.01

 enfans

-0.94

 Majefty

-0.94

 Monfieur

-0.94

 igång

-0.93

 ſtate

-0.93

POSITIVE LOGITS

mal

1.70

Mal

1.70

MAL

1.65

Mal

1.64

mal

1.49

MAL

1.39

 Malone

1.37

 Maloney

1.20

 Mald

1.08

 Malden

1.04

Activations Density 0.035%

terms related to "mal" or "malignant" conditions

No Comments

No Known Activations